INDEX

Explanations

markers indicating a list or enumeration

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amedi

-0.16

£½

-0.16

VERTISEMENT

-0.14

owie

-0.14

ains

-0.14

zel

-0.14

alet

-0.14

æĪ

-0.13

oppins

-0.13

erse

-0.13

POSITIVE LOGITS

rost

0.19

æī¬

0.15

dater

0.15

 rhet

0.14

LLL

0.14

istrovstvÃŃ

0.14

Borders

0.13

Å

0.13

Peripheral

0.13

 uncomment

0.13

Activations Density 0.004%

markers indicating a list or enumeration

No Comments

No Known Activations